水利工程混凝土钢筋网绑扎率的视觉智能评价方法  

Visual Intelligent Evaluation Method for Binding Rate of Concrete Reinforcement Mesh in Hydraulic Engineering

作　　者：庞小荣 刘全[1] 王浩[2] 张宏阳 孟歆 周剑 PANG Xiaorong;LIU Quan;WANG Hao;ZHANG Hongyang;MENG Xin;ZHOU Jian(State Key Laboratory of Water Resources Engineering and Management,Wuhan University,Wuhan 430072,Hubei,China;Institute of Water Engineering Sciences,Wuhan University,Wuhan 430072,Hubei,China;Changjiang Survey,Planning,Design and Research Co.,Ltd.,Wuhan 430010,Hubei,China)

机构地区：[1]武汉大学水资源工程与调度全国重点实验室,湖北武汉430072 [2]武汉大学水工程科学研究院,湖北武汉430072 [3]长江勘测规划设计研究有限责任公司,湖北武汉430010

出　　处：《水力发电》2024年第10期59-64,共6页Water Power

摘　　要：水利水电工程大体积混凝土常采用间隔绑扎的单层钢筋网。针对当前钢筋网绑扎率人工目视验收存在效率慢、精度低的问题,基于计算机视觉与拓扑分析提出了钢筋网绑扎率的视觉智能评价方法。训练了识别钢筋网图像中钢筋交叉点与绑扎状态的目标检测模型;针对无序的识别框坐标,提出“十字交叉”方法与生长编号算法,实现了钢筋网拓扑重构;根据交叉点拓扑编号提出基于节点自由度的最优绑扎方法。实验分析表明,方法解决了人工定性评价的缺陷,兼具精度与效率优势,实现了钢筋网绑扎率的视觉智能评价,能够为水利工程智能建造提供有力支持。The mass concrete of hydropower projects often adopts single-layer reinforcing mesh with interval binding.Aiming at the problems of slow efficiency and low accuracy in manual visual acceptance of reinforcing mesh binding rate,a visual intelligent evaluation method for reinforcing mesh binding rate is proposed based on computer vision and topology analysis.The target detection model for identifying the intersection and binding state of steel bars in the reinforcing mesh image is trained.Aiming at the disordered identification frame coordinates,the“Cross”method and the growth numbering algorithm are proposed to realize the topology reconstruction of the reinforcing mesh.According to the intersection topology number,an optimal binding method based on node degree of freedom is proposed.The experimental analysis shows that the method solves the defects of manual qualitative evaluation,has the advantages of accuracy and efficiency,and realizes the visual intelligent evaluation of the binding rate of reinforcing mesh,which can provide strong support for the intelligent construction of water conservancy projects.

关 键 词：智能建造 混凝土备仓 钢筋网绑扎 计算机视觉 拓扑分析 模式识别 

分 类 号：TV523[水利工程—水利水电工程]